Classification Prediction Model of Students' College English Scores Based on Online Course Learning Data

In the process of information education construction, online course teaching, as an important application of information technology in the field of education, has gradually become an important part of college English education. However, in the face of changeable online learning forms and numerous student groups, the limitations of traditional teaching management methods are increasingly prominent, which in turn has a negative impact on the overall teaching effect. In this regard, based on the current situation of college English teaching in colleges and universities, this paper will deeply analyze the feasibility of the application of artificial intelligence and deep learning technology in this field, and put forward a set of performance classification prediction model driven by students' learning data, aiming at identifying students' learning risks conveniently and efficiently and improving the effectiveness of teaching management. Practice has proved that the whole model consists of two branch neural networks, one is the learning behavior feature extraction network with Bi-LSTM as the core; The second is the student attribute feature extraction network with DNN as the core. The two-part network completes the construction of joint feature vector under the fusion mechanism, and completes the prediction of students' grades accordingly, thus providing a more scientific and objective basis for teaching management.